As the flood of critical biological information swells, medical researchers will drown in it unless it is creatively and effectively organized. Currently, Protein Reviews on the Web (PROW) is a pilot project to use human proteins as the organizing principle around which to build a conceptual web of biological concepts. Its success depends on our developing excellent strategies for organizing information and our tapping into the expertise of the biologic community to help develop and sustain such a resource. The end-product is a WWW resource/journal (http://www.ncbi.nlm.nih.gov/prow) consisting of online authoritative reviews on human proteins (and their orthologs in other species) maintained by biological experts in the research community. One major advance this year has been the introduction of a second-generation format for the CD guides. This format is more comprehensive than our first-generation format with respect to the range of information in its various sections. Of particular importance in the new format is the addition of an abstract and a summary sentence, which provide a concise overview that is useful to a wide variety of readers. A second major accomplishment this year is an ongoing collaboration with the 7th International Workshop on Leukocyte Differentiation Antigens; PROW is commissioning Guides on most of the 60 approximately molecules which were studied and given CD designations this year. In the past we tended to rely on expertise of active Workshop participants; now we are aggressively soliciting contributions from world experts regardless of their workshop participation. The acceptance rate for authorship and reviewing has been high, apparently reflecting the growing credibility of PROW (and the Workshops). The utilization of PROW is growing rapidly, with peak utilization of 30,000 hits/month. We are making progress in developing the novel KBTool database design and efficient user interface to store and retrieve the complex web of information associated with biological entities and their associated research community. This database is the infrastructure on which PROW is based, enabling biological information to be viewed as an integrated fabric of information. Search and categorization components have been improved. Furthermore, we have developed an XML file format for export of PROW Guides from the underlying KBTool database; guides in different formats are then generated using XSL transformation. This will make the data readily available to external data consumers.